CN106199338A - A kind of discrimination method of short trouble type voltage sag source - Google Patents

Abstract

The invention discloses the discrimination method of a kind of short trouble type voltage sag source, including: step 10) gathers short trouble type voltage dip sample data, after described sample data is carried out data prediction, forms coupling database data；Step 20) obtain the short trouble type voltage dip Wave data surveyed, after standardization, as data to be matched；Step 30) calculate dynamic time warping distance: coupling database data and data to be matched are normalized, calculate the data to be matched after normalization and the dynamic time warping distance mating database data；Step 40) identify short trouble type: according to dynamic time warping distance size, it is determined that short-circuit-type and the short circuit of the voltage dip of actual measurement are separate, it is achieved the identification of voltage sag source.The voltage dip that short trouble can be caused by the method carries out accurate recognition, it is adaptable to engineering is actual.

Description

A kind of discrimination method of short trouble type voltage sag source

Technical field

The invention belongs to electric energy quality monitoring and analysis technical field, it particularly relates to a kind of short trouble type voltage The discrimination method in source drops temporarily.

Background technology

Along with constantly building and development of electrical network, all types of industries power consumer accesses electrical network.These users have employed advanced person Automatization, intellectuality, digitized voltage dip sensitive equipment, such as PLC, converter, bus, catalyst, relay, control Devices etc., propose the highest requirement to supply voltage quality.Once these equipment quit work because of voltage dip, complete equipment Or streamline all can be affected, bring huge loss to enterprise.Manufacturing enterprise generally believes that voltage dip has risen For most important power quality problem, being the main interference of modern industry power load normal safe work, each side is given the most To pay close attention to.

Voltage dip has become as the power quality problem needing solution badly.Voltage dip control measures mainly includes carrying at present The tolerance of high equipment, the minimizing generation of voltage sag source, the propagation in disturbance suppression source and configuration voltage sag compensation device. And accurate recognition voltage sag source is the premise administering voltage dip.The identification of voltage sag source contributes to selecting reasonably to administer Measure, can be that the coordination dispute between supply of electric power department and user provides foundation simultaneously.

The identification of voltage sag source refers to according to voltage dip its producing cause of data characteristics identification detected.Both at home and abroad The discrimination method to voltage sag source has carried out numerous studies, and obtains numerous achievements, mainly have wavelet transformation, S-transformation, The methods such as vector machine.Most voltage dip in power system is by caused by short trouble, and voltage dip is through inhomogeneity Type transformator drops type temporarily it can also happen that change after propagating.Above-mentioned discrimination method is merely able to effectively distinguish induction conductivity and opens Voltage dip dynamic, that no-load transformer is excitatory and short trouble these three reason causes, but the voltage causing short trouble is temporary Fall cannot pick out short trouble type concrete in system and short circuit phase further, and the method which greatly limits is at Practical Project In application.

Summary of the invention

Technical problem: the technical problem to be solved is: distinguishing of a kind of short trouble type voltage sag source is provided Knowledge method, the voltage dip that short trouble can be caused by the method carries out accurate recognition, it is adaptable to engineering is actual, for engineering people Member's positioning voltage drops source temporarily, analyzes the voltage dip mechanism of transmission and select the offer of voltage dip control measures effectively to help.

Technical scheme: for solving above-mentioned technical problem, the embodiment of the present invention uses a kind of short trouble type voltage sag source Discrimination method, the method comprises the following steps:

Step 10) gather short trouble type voltage dip sample data, after described sample data is carried out data prediction, Form coupling database data；

Step 20) obtain the short trouble type voltage dip Wave data of actual measurement, after standardization, as treating Join data；

Step 30) calculate dynamic time warping distance: being normalized coupling database data and data to be matched, calculating is returned Data to be matched after one change and the dynamic time warping distance mating database data；

Step 40) identify short trouble type: according to dynamic time warping distance size, it is determined that the voltage dip of actual measurement Short-circuit-type and short circuit are separate, it is achieved the identification of voltage sag source.

As preference, described step 10) specifically include:

Step 101) gather short circuit failed-type voltage dip sample data in historical data, described sample data includes abc The voltage dip ABC phase data that phase ground short circuit, ab phase ground short circuit and three kinds of short troubles of a phase ground short circuit cause, often group Sample data is arranged in data sequence as the following formula:

U_A(1),U_B(1),U_C(1)

U_A(2),U_B(2),U_C(2)

U_A(M),U_B(M),U_C(M)

In formula, U_A(1) the A phase voltage in the 1st group of sample data, U are represented_B(1) represent that the B phase in the 1st group of sample data is electric Pressure, U_C(1) the C phase voltage in the 1st group of sample data, U are represented_A(2) the A phase voltage in the 2nd group of sample data, U are represented_B(2) table Show the B phase voltage in the 2nd group of sample data, U_C(2) the C phase voltage in the 2nd group of sample data, U are represented_A(M) M group sample is represented A phase voltage in notebook data, U_B(M) the B phase voltage in M group sample data, U are represented_C(M) represent in M group sample data C phase voltage, M represents the quantity of sample；

Step 102) to step 101) data sequence that gathers carries out data prediction, often group data sequence is arranged in 1 row, The data phase sequence sample data as bc, ca phase ground short circuit is adjusted for ab phase ground short circuit, for a phase ground short circuit, adjusts Entire data phase sequence, as the sample data of b, c phase ground short circuit, forms the coupling database data of 7 row 3M row, mates database data form As follows:

U_A(1),U_A(2)…U_A(M),U_B(1),U_B(2)…U_B(M),U_C(1),U_C(2)…U_C(M)}①

In formula, it 1. it is the coupling database data of abc phase ground short circuit；2. it is the coupling database data of ab phase ground short circuit；3. it is a The coupling database data of phase ground short circuit；First behavior abc phase ground short circuit, the second behavior ab phase ground short circuit, the third line is bc Phase ground short circuit, fourth line is ca phase ground short circuit, and fifth line is a phase ground short circuit, the 6th behavior b phase ground short circuit, the 7th Behavior c phase ground short circuit.

As preference, described step 20) including: obtain the voltage dip Wave data of actual measurement at points of common connection, Form is as follows:

$\begin{array}{c}{\stackrel{\‾}{U}}_A\left(1\right),{\stackrel{\‾}{U}}_B\left(1\right),{\stackrel{\‾}{U}}_C\left(1\right)\\ {\stackrel{\‾}{U}}_A\left(2\right),{\stackrel{\‾}{U}}_B\left(2\right),{\stackrel{\‾}{U}}_C\left(2\right)\\ .\\ .\\ .\\ {\stackrel{\‾}{U}}_A\left(N\right),{\stackrel{\‾}{U}}_B\left(N\right),{\stackrel{\‾}{U}}_C\left(N\right)\end{array}$

In formula,Represent the A phase voltage of the 1st group of data in measured data,Represent the 1st group of number in measured data According to B phase voltage,Represent the C phase voltage of the 1st group of data in measured data,Represent the 2nd group of number in measured data According to A phase voltage,Represent the B phase voltage of the 2nd group of data in measured data,Represent the 2nd group of number in measured data According to C phase voltage,Represent the A phase voltage of N group data in measured data,Represent N group in measured data The B phase voltage of data,Represent the C phase voltage of N group data in measured data；

It is standardized processing to the actual measurement voltage dip Wave data obtained, data are lined up the to be matched of 1 row 3N row Data, form is as follows:

${\stackrel{\‾}{U}}_A\left(1\right),{\stackrel{\‾}{U}}_A\left(2\right)...{\stackrel{\‾}{U}}_A\left(N\right),{\stackrel{\‾}{U}}_B\left(1\right),{\stackrel{\‾}{U}}_B\left(2\right)...{\stackrel{\‾}{U}}_B\left(N\right),{\stackrel{\‾}{U}}_C\left(1\right),{\stackrel{\‾}{U}}_C\left(2\right)...{\stackrel{\‾}{U}}_C\left(N\right).$

As preference, described step 30) including:

Step 301) often row coupling database data and data to be matched are normalized, difference reduction to interval [-1, 1], in, normalized function is:

$U^{\′}\left(p\right)=\frac{2U\left(p\right)-U_{min}-U_{max}}{U_{max}-U_{min}},(p=1,2,3...,3M)$

${\stackrel{\‾}{U}}^{\′}\left(q\right)=\frac{2\stackrel{\‾}{U}\left(q\right)-{\stackrel{\‾}{U}}_{min}-{\stackrel{\‾}{U}}_{max}}{{\stackrel{\‾}{U}}_{\max }-{\stackrel{\‾}{U}}_{\min }},(q=1,2,3...,3N)$

In formula, U represents the coupling database data a line sequence before normalization,Represent the data to be matched before normalization A line sequence；U_maxRepresent the maximum of U,RepresentMaximum；U_minRepresent the minima of U,Represent's Minima；U' mates database data a line sequence after representing normalization,Represent a line sequence of the data to be matched after normalization；

Step 302) calculate the data to be matched after normalization and the dynamic time mated between database data after normalization Deflection distance:

First, utilize U' andConstruct the distance matrix A of 3M row 3N row, be:

In formula, element in ARepresent the data to be matched after normalizationWith mate storehouse The alignment distance of data U '；

U' andDynamic time warping distanceFor:

$\left\{\begin{array}{c}DTW(U^{\′},{\stackrel{\‾}{U}}^{\′})=D(3M,3N)\\ D(0,0)=0\\ D(0,\mathrm{\∞})=0\\ D(\mathrm{\∞},0)=0\\ D(i,j)=a_{ij}+min\left\{D\right(i-1,j-1),D(i-1,j),D(i,j-1\left)\right\}\\ 1\≤i\≤3M,1\≤j\≤3N\end{array}\right.$

In formula, (i j) represents a to D_ijMinimum dynamic time warping distance sum with its leading portion；a_ijRepresent element d in A (x_i,x_j)；∞ represents any number being numbered 1 to 3M or 1 to 3N；

Step 303) coupling database data line down, repeat step 302), until having calculated data to be matched and all row The dynamic time warping distance of coupling database data, obtains 7 dynamic time warping distances.

As preference, described step 40) including: by a line coupling storehouse number minimum for 7 dynamic time warping distances According to as the data the highest with Data Matching degree to be matched, data to be matched belong to same class with mating the row data in database data Type voltage dip.

Beneficial effect: compared with prior art, the embodiment of the present invention has the advantages that the method can accurately be distinguished Know short trouble type voltage sag source type and short circuit separate.Traditional discrimination method is merely able to effectively distinguish induction electric The voltage dip that machine starts, no-load transformer is excitatory and short trouble these three reason causes, but the electricity that short trouble is caused Pressure fall temporarily cannot pick out short trouble type concrete in system further and short circuit is separate.The method of the present embodiment can profit Being analyzed identification by the voltage dip data of actual measurement in engineering, method is the most effective, it is easy to engineer applied.

Accompanying drawing explanation

Fig. 1 is the flow chart of the embodiment of the present invention.

The voltage dip sample data oscillogram gathered when Fig. 2 (a) is abc phase ground short circuit in the embodiment of the present invention.

The voltage dip sample data oscillogram gathered when Fig. 2 (b) is ab phase ground short circuit in the embodiment of the present invention.

The voltage dip sample data oscillogram gathered when Fig. 2 (c) is a phase ground short circuit in the embodiment of the present invention.

Fig. 3 is voltage dip measured data oscillogram in the embodiment of the present invention.

Detailed description of the invention

Below in conjunction with example and accompanying drawing, the technical scheme of the embodiment of the present invention is described further.

As it is shown in figure 1, the embodiment of the present invention provides the discrimination method of a kind of short trouble type voltage sag source, including following Step:

Step 10) gather short trouble type voltage dip sample data, after described sample data is carried out data prediction, Form coupling database data；

Step 20) obtain the short trouble type voltage dip Wave data surveyed, after standardization,

As data to be matched；

Step 30) calculate dynamic time warping distance: being normalized coupling database data and data to be matched, calculating is returned Data to be matched after one change and the dynamic time warping distance mating database data；

Step 40) identify short trouble type: according to dynamic time warping distance size, it is determined that the voltage dip of actual measurement Short-circuit-type and short circuit are separate, it is achieved the identification of voltage sag source.

In the above-described embodiments, described step 10) specifically include:

Step 101) gather short circuit failed-type voltage dip sample data in historical data, described sample data includes abc The voltage dip ABC phase data that phase ground short circuit, ab phase ground short circuit and three kinds of short troubles of a phase ground short circuit cause, often group Sample data is arranged in data sequence as the following formula:

U_A(1),U_B(1),U_C(1)

U_A(2),U_B(2),U_C(2)

U_A(M),U_B(M),U_C(M)

In formula, U_A(1) the A phase voltage in the 1st group of sample data, U are represented_B(1) represent that the B phase in the 1st group of sample data is electric Pressure, U_C(1) the C phase voltage in the 1st group of sample data, U are represented_A(2) the A phase voltage in the 2nd group of sample data, U are represented_B(2) table Show the B phase voltage in the 2nd group of sample data, U_C(2) the C phase voltage in the 2nd group of sample data, U are represented_A(M) M group sample is represented A phase voltage in notebook data, U_B(M) the B phase voltage in M group sample data, U are represented_C(M) represent in M group sample data C phase voltage, M represents the quantity of sample；

Step 102) to step 101) data sequence that gathers carries out data prediction, often group data sequence is arranged in 1 row, The data phase sequence sample data as bc, ca phase ground short circuit is adjusted for ab phase ground short circuit, for a phase ground short circuit, adjusts Entire data phase sequence, as the sample data of b, c phase ground short circuit, forms the coupling database data of 7 row 3M row, mates database data form As follows:

U_A(1),U_A(2)…U_A(M),U_B(1),U_B(2)…U_B(M),U_C(1),U_C(2)…U_C(M)}①

In formula, it 1. it is the coupling database data of abc phase ground short circuit；2. it is the coupling database data of ab phase ground short circuit；3. it is a The coupling database data of phase ground short circuit；First behavior abc phase ground short circuit, the second behavior ab phase ground short circuit, the third line is bc Phase ground short circuit, fourth line is ca phase ground short circuit, and fifth line is a phase ground short circuit, the 6th behavior b phase ground short circuit, the 7th Behavior c phase ground short circuit.

In the above-described embodiments, described step 20) specifically include: obtain points of common connection (corresponding English full name Point of Common Coupling；Referred to as PCC point in literary composition) the voltage dip Wave data of place's actual measurement, form is as follows:

$\begin{array}{c}{\stackrel{\‾}{U}}_A\left(1\right),{\stackrel{\‾}{U}}_B\left(1\right),{\stackrel{\‾}{U}}_C\left(1\right)\\ {\stackrel{\‾}{U}}_A\left(2\right),{\stackrel{\‾}{U}}_B\left(2\right),{\stackrel{\‾}{U}}_C\left(2\right)\\ .\\ .\\ .\\ {\stackrel{\‾}{U}}_A\left(N\right),{\stackrel{\‾}{U}}_B\left(N\right),{\stackrel{\‾}{U}}_C\left(N\right)\end{array}$

In formula,Represent the A phase voltage of the 1st group of data in measured data,Represent the 1st group of number in measured data According to B phase voltage,Represent the C phase voltage of the 1st group of data in measured data,Represent the 2nd group of number in measured data According to A phase voltage,Represent the B phase voltage of the 2nd group of data in measured data,Represent the 2nd group of number in measured data According to C phase voltage,Represent the A phase voltage of N group data in measured data,Represent N group in measured data The B phase voltage of data,Represent the C phase voltage of N group data in measured data；

It is standardized processing to the actual measurement voltage dip Wave data obtained, data are lined up the to be matched of 1 row 3N row Data, form is as follows:

${\stackrel{\‾}{U}}_A\left(1\right),{\stackrel{\‾}{U}}_A\left(2\right)...{\stackrel{\‾}{U}}_A\left(N\right),{\stackrel{\‾}{U}}_B\left(1\right),{\stackrel{\‾}{U}}_B\left(2\right)...{\stackrel{\‾}{U}}_B\left(N\right),{\stackrel{\‾}{U}}_C\left(1\right),{\stackrel{\‾}{U}}_C\left(2\right)...{\stackrel{\‾}{U}}_C\left(N\right).$

In the above-described embodiments, described step 30) specifically include:

Step 301) often row coupling database data and data to be matched are normalized, difference reduction to interval [-1, 1], in, normalized function is:

$U^{\′}\left(p\right)=\frac{2U\left(p\right)-U_{min}-U_{max}}{U_{max}-U_{min}},(p=1,2,3...,3M)$

${\stackrel{\‾}{U}}^{\′}\left(q\right)=\frac{2\stackrel{\‾}{U}\left(q\right)-{\stackrel{\‾}{U}}_{min}-{\stackrel{\‾}{U}}_{max}}{{\stackrel{\‾}{U}}_{\max }-{\stackrel{\‾}{U}}_{\min }},(q=1,2,3...,3N)$

In formula, U represents the coupling database data a line sequence before normalization,Represent the data to be matched before normalization A line sequence；U_maxRepresent the maximum of U,RepresentMaximum；U_minRepresent the minima of U,Represent's Minima；U' mates database data a line sequence after representing normalization,Represent a line sequence of the data to be matched after normalization；

Step 302) calculate the data to be matched after normalization and the dynamic time mated between database data after normalization Deflection distance:

First, utilize U' andConstruct the distance matrix A of 3M row 3N row, be:

In formula, element in ARepresent the data to be matched after normalizationWith mate storehouse The alignment distance of data U '；

U' andDynamic time warping distanceFor:

$\left\{\begin{array}{c}DTW(U^{\′},{\stackrel{\‾}{U}}^{\′})=D(3M,3N)\\ D(0,0)=0\\ D(0,\mathrm{\∞})=0\\ D(\mathrm{\∞},0)=0\\ D(i,j)=a_{ij}+min\left\{D\right(i-1,j-1),D(i-1,j),D(i,j-1\left)\right\}\\ 1\≤i\≤3M,1\≤j\≤3N\end{array}\right.$

In formula, (i j) represents a to D_ijMinimum dynamic time warping distance sum with its leading portion；a_ijRepresent element d in A (x_i,x_j)；∞ represents any number being numbered 1 to 3M or 1 to 3N；

Step 303) coupling database data line down, repeat step 302), until having calculated data to be matched and all row The dynamic time warping distance of coupling database data, obtains 7 dynamic time warping distances.

In the above-described embodiments, described step 40) specifically include: a line minimum by 7 dynamic time warping distances Coupling database data is as the data the highest with Data Matching degree to be matched, and data to be matched belong to mating the row data in database data In same type voltage dip.Identification result is as follows:

The discrimination method of the embodiment of the present invention, utilize dynamic time warping distance weigh actual measurement voltage dip data with Join the similarity of database data, search out Wave data most like with data to be matched in coupling storehouse, and then judge the electricity of actual measurement Type and the short circuit of pressure fall temporarily are separate.In prior art, the identification of voltage sag source is concentrated mainly on differentiation induction conductivity and opens Voltage dip dynamic, that no-load transformer is excitatory and short trouble these three reason causes, and in power system, short trouble is electricity The main cause of pressure fall temporarily, it is therefore desirable to pick out short trouble type concrete in system further and short circuit is separate.This reality Execute the voltage dip that short trouble can cause by example method and carry out accurate recognition, it is adaptable to engineering is actual, fixed for engineering staff Position voltage sag source, the analysis voltage dip mechanism of transmission and selection voltage dip control measures provide and effectively help.

Enumerate a specific embodiment below.

As a example by certain 10kV bus, the 380V user side under bus arranges voltage measuring apparatus and carries out voltage dip number According to collection.Emulation sets gradually abc phase ground short circuit, ab phase ground short circuit and a phase ground short circuit and forms voltage dip sample Waveform, a length of 100ms during short circuit；Bc phase ground short circuit is set and forms voltage dip measured waveform, a length of 50ms during short circuit；Data Sample frequency is 10kHz.

Perform the discrimination method of the embodiment of the present invention, comprise the following steps:

Step 10) gather short trouble type voltage dip sample data, including abc phase ground short circuit, ab phase ground short circuit The voltage dip ABC phase data caused with 3 kinds of short troubles of a phase ground short circuit, forms the coupling of 7 row 3000 row through pretreatment Database data.The voltage dip sample data of this example collection as in figure 2 it is shown, in figure abscissa represent sampling number, vertical coordinate table Show the perunit value of voltage magnitude.The voltage dip sample data figure gathered when Fig. 2 (a) is abc phase ground short circuit；Fig. 2 (b) is ab The voltage dip sample data figure gathered during phase ground short circuit；The voltage dip sample gathered when Fig. 2 (c) is a phase ground short circuit Datagram.

Step 20) obtain the voltage dip Wave data of actual measurement of PCC point, form after standardization that 1 row 3000 arranges treats Join data.The actual measurement voltage dip Wave data of this case-based system as it is shown on figure 3, in figure abscissa represent sampling number, vertical sit Mark represents the perunit value of voltage magnitude.

Step 30) calculate data to be matched and the DTW distance mating the every a line of database data after normalization, obtain 7 altogether DTW distance, result of calculation is as shown in table 1.

Table 1 DTW distance result of calculation

Numbering	1	2	3	4	5	6	7
								DTW distance	653.11	1411.70	5.17	756.86	1330.30	579.38	167.16

Step 40) data to be matched are the highest with the 3rd row data matching degree mating database data, and both belong to same type electricity Pressure fall temporarily, therefore the voltage dip that identification result is actual measurement is caused by bc phase ground short circuit fault.The identification result of this enforcement Consistent with practical situation, identification result is accurate.

The ultimate principle of the present invention, principal character and advantage have more than been shown and described.Those skilled in the art should Understand, the present invention is not limited by above-mentioned specific embodiment, the description in above-mentioned specific embodiment and description be intended merely to into One step explanation the present invention principle, without departing from the spirit and scope of the present invention, the present invention also have various change and Improving, these changes and improvements both fall within scope of the claimed invention.The scope of protection of present invention is wanted by right Book and equivalent thereof is asked to define.

Claims (5)

1. the discrimination method of a short trouble type voltage sag source, it is characterised in that the method comprises the following steps:

Step 10) gather short trouble type voltage dip sample data, after described sample data is carried out data prediction, formed Coupling database data；

Step 20) obtain the short trouble type voltage dip Wave data surveyed, after standardization, as number to be matched According to；

Step 30) calculate dynamic time warping distance: coupling database data and data to be matched are normalized, calculate normalization After data to be matched and the dynamic time warping distance mating database data；

Step 40) identify short trouble type: according to dynamic time warping distance size, it is determined that the short circuit of the voltage dip of actual measurement Type and short circuit are separate, it is achieved the identification of voltage sag source.

The discrimination method of short trouble type voltage sag source the most according to claim 1, it is characterised in that described step 10) specifically include:

Step 101) gather short circuit failed-type voltage dip sample data in historical data, described sample data includes that abc connects The voltage dip ABC phase data that ground short circuit, ab phase ground short circuit and three kinds of short troubles of a phase ground short circuit cause, often organizes sample Data are arranged in data sequence as the following formula:

$\begin{array}{c}{U}_A\left(1\right),U_B\left(1\right),U_C\left(1\right)\\ U_A\left(2\right),U_B\left(2\right),U_C\left(2\right)\\ \·\\ \·\\ \·\\ U_A\left(M\right),U_B\left(M\right),U_C\left(M\right)\end{array}$

In formula, U_A(1) the A phase voltage in the 1st group of sample data, U are represented_B(1) the B phase voltage in the 1st group of sample data is represented, U_C(1) the C phase voltage in the 1st group of sample data, U are represented_A(2) the A phase voltage in the 2nd group of sample data, U are represented_B(2) represent B phase voltage in 2nd group of sample data, U_C(2) the C phase voltage in the 2nd group of sample data, U are represented_A(M) M group sample is represented A phase voltage in data, U_B(M) the B phase voltage in M group sample data, U are represented_C(M) C in M group sample data is represented Phase voltage, M represents the quantity of sample；

Step 102) to step 101) data sequence that gathers carries out data prediction, often group data sequence is arranged in 1 row, for Ab phase ground short circuit adjusts the data phase sequence sample data as bc, ca phase ground short circuit, for a phase ground short circuit, adjusts number According to phase sequence as the sample data of b, c phase ground short circuit, forming the coupling database data of 7 row 3M row, coupling database data form is as follows:

U_A(1),U_A(2)…U_A(M),U_B(1),U_B(2)…U_B(M),U_C(1),U_C(2)…U_C(M)}①

In formula, it 1. it is the coupling database data of abc phase ground short circuit；2. it is the coupling database data of ab phase ground short circuit；3. connect for a The coupling database data of ground short circuit；First behavior abc phase ground short circuit, the second behavior ab phase ground short circuit, the third line is that bc connects Ground short circuit, fourth line is ca phase ground short circuit, and fifth line is a phase ground short circuit, the 6th behavior b phase ground short circuit, the 7th behavior c Phase ground short circuit.

The discrimination method of short trouble type voltage sag source the most according to claim 2, it is characterised in that described step 20) including: obtaining the voltage dip Wave data of actual measurement at points of common connection, form is as follows:

$\begin{array}{c}{\stackrel{\‾}{U}}_A\left(1\right),{\stackrel{\‾}{U}}_B\left(1\right),{\stackrel{\‾}{U}}_C\left(1\right)\\ {\stackrel{\‾}{U}}_A\left(2\right),{\stackrel{\‾}{U}}_B\left(2\right),{\stackrel{\‾}{U}}_C\left(2\right)\\ \begin{array}{c}.\\ .\\ .\end{array}\\ {\stackrel{\‾}{U}}_A\left(N\right),{\stackrel{\‾}{U}}_B\left(N\right),{\stackrel{\‾}{U}}_C\left(N\right)\end{array}$

In formula,Represent the A phase voltage of the 1st group of data in measured data,Represent the 1st group of data in measured data B phase voltage,Represent the C phase voltage of the 1st group of data in measured data,Represent the 2nd group of data in measured data A phase voltage,Represent the B phase voltage of the 2nd group of data in measured data,Represent the 2nd group of data in measured data C phase voltage,Represent the A phase voltage of N group data in measured data,Represent N group data in measured data B phase voltage,Represent the C phase voltage of N group data in measured data；

It is standardized processing to the actual measurement voltage dip Wave data obtained, data is lined up the data to be matched of 1 row 3N row, Form is as follows:

${\stackrel{\‾}{U}}_A\left(1\right),{\stackrel{\‾}{U}}_A\left(2\right)...{\stackrel{\‾}{U}}_A\left(N\right),{\stackrel{\‾}{U}}_B\left(1\right),{\stackrel{\‾}{U}}_B\left(2\right)...{\stackrel{\‾}{U}}_B\left(N\right),{\stackrel{\‾}{U}}_C\left(1\right),{\stackrel{\‾}{U}}_C\left(2\right)...{\stackrel{\‾}{U}}_C\left(N\right).$

The discrimination method of short trouble type voltage sag source the most according to claim 3, it is characterised in that described step 30) including:

Step 301) often row coupling database data and data to be matched are normalized, reduction is to interval [-1,1] respectively In, normalized function is:

$U^{\′}\left(p\right)=\frac{2U\left(p\right)-U_{min}-U_{max}}{U_{max}-U_{min}},(p=1,2,3...,3M)$

${\stackrel{\‾}{U}}^{\′}\left(q\right)=\frac{2\stackrel{\‾}{U}\left(q\right)-{\stackrel{\‾}{U}}_{min}-{\stackrel{\‾}{U}}_{max}}{{\stackrel{\‾}{U}}_{\max }-{\stackrel{\‾}{U}}_{\min }},(q=1,2,3...,3N)$

In formula, U represents the coupling database data a line sequence before normalization,One of data to be matched before expression normalization Line order arranges；U_maxRepresent the maximum of U,RepresentMaximum；U_minRepresent the minima of U,RepresentMinimum Value；U' mates database data a line sequence after representing normalization,Represent a line sequence of the data to be matched after normalization；

Step 302) calculate the data to be matched after normalization and the dynamic time warping mated between database data after normalization Distance:

First, utilize U' andConstruct the distance matrix A of 3M row 3N row, be:

In formula, element in ARepresent the data to be matched after normalizationWith mate database data The alignment distance of U '；

U' andDynamic time warping distanceFor:

$\left\{\begin{array}{c}DTW(U^{\′},{\stackrel{\‾}{U}}^{\′})=D(3M,3N)\\ D(0,0)=0\\ D(0,\mathrm{\∞})=0\\ D(\mathrm{\∞},0)=0\\ D(i,j)=a_{ij}+min\{D(i-1,j-1),D(i-1,j),D(i,j-1)\}\\ 1\≤i\≤3M,1\≤j\≤3N\end{array}\right.$

In formula, (i j) represents a to D_ijMinimum dynamic time warping distance sum with its leading portion；a_ijRepresent element d (x in A_i, x_j)；∞ represents any number being numbered 1 to 3M or 1 to 3N；

Step 303) coupling database data line down, repeat step 302), mate until having calculated data to be matched with all row The dynamic time warping distance of database data, obtains 7 dynamic time warping distances.

The discrimination method of short trouble type voltage sag source the most according to claim 4, it is characterised in that described step 40) including: using a line coupling database data minimum for 7 dynamic time warping distances as the highest with Data Matching degree to be matched Data, data to be matched belong to same type voltage dip with mating the row data in database data.